Google, Microsoft, and OpenAI Back Linux Foundation's Appia AI Standards Initiative (nerds.xyz) 24
BrianFagioli writes: Google, Microsoft, OpenAI, Arm, Mastercard, Siemens, and other companies have joined the newly launched Appia Foundation under the Linux Foundation. The project aims to create common specifications and assessment frameworks that organizations can use to demonstrate AI systems meet emerging safety, trust, and compliance requirements. According to the Linux Foundation, the framework is designed to allow conformity evidence to be reused across the AI supply chain, potentially reducing duplicate assessments and compliance costs. The announcement comes as governments around the world move toward enforcing AI regulations and organizations face increasing pressure to prove AI systems are trustworthy. "As international standards and legal frameworks become more established, global organizations need a consistent, practical way to verify that AI systems conform to new expectations," said Jim Zemlin, CEO of the Linux Foundation. "The Appia Foundation establishes a neutrally governed environment where the entire industry can collaborate on a common assessment framework. By building this infrastructure in the open, we are helping organizations reduce complexity, lower operational costs and build trust."
Craig Shank, Executive Director of the Appia Foundation, added: "AI systems now make decisions about people's loans, their children's schools and their jobs. People on the receiving end deserve to know those systems were built and assessed against criteria that hold up to scrutiny. The Appia Foundation was formed to do that work: creating publicly available specifications that organizations across the AI value chain use to demonstrate their systems meet those criteria. By establishing this open framework, we are building the accountability layer required to scale safe and trusted AI across major industries."
Craig Shank, Executive Director of the Appia Foundation, added: "AI systems now make decisions about people's loans, their children's schools and their jobs. People on the receiving end deserve to know those systems were built and assessed against criteria that hold up to scrutiny. The Appia Foundation was formed to do that work: creating publicly available specifications that organizations across the AI value chain use to demonstrate their systems meet those criteria. By establishing this open framework, we are building the accountability layer required to scale safe and trusted AI across major industries."
"AI systems now make decisions" (Score:1)
Re: (Score:1)
there is no Intelligence in today's AI
So a lot like people then.
Re: (Score:1)
Re: (Score:2)
current AI has no ability to reason or understand, they are just pattern matching algorithms.
I'm not sure about AI, but human stupidity is very real. And it becomes evident in the slashdot comments for any topic that raises emotions. The rich guy, the orange guy, and AI being prime examples.
Given the rather obvious reasoning of reasoning models, you have to wonder what "no ability to reason" intends to mean. i'm sure bloodhawk isn't a moron, or deliberately trying to sound like a moron, but rather expressing some sort of emotion. I guess "AI reasoning is not the same as human reasoning"? Then w
Re: (Score:2)
A. Coward, you keep saying that, but it does not mean anything. You show no understanding, just repeating words you think fit.
So how about defining some terms then? What is understanding, if we are going to try to apply it to non-human intelligence?
For decades we have been applying these questions to animals. Mirror self-recognition, theory of mind etc.
The problem is that most people have really actually thought about it, so do not have a meaningful definition of "understanding". Yet many of them make c
Re: (Score:2)
So why don't you try saying what it actually is, instead of "no it isn't"? Why hide under AC if not actually saying anything?
There is no point saying "its not real", if you can't define what real is, especially in terms of a testable theory. they are empty, meaningless words. Ironically, an AI could show much deeper understanding than the AC - based on observable outputs. (ie empirical evidence)
Re: (Score:2)
What exactly are you asking for here?
If you haven't met the burden of proof required to convince someone of your correctness, why are they obliged to provide you with a more correct answer?
In reality, they only need to provide a counterexample to your assertions.
Re: (Score:1)
A. Coward, you keep saying that, but it does not mean anything. You show no understanding, just repeating words you think fit.
So how about defining some terms then? What is understanding, if we are going to try to apply it to non-human intelligence? For decades we have been applying these questions to animals. Mirror self-recognition, theory of mind etc.
The problem is that most people have really actually thought about it, so do not have a meaningful definition of "understanding". Yet many of them make confident pronouncements anyway. I call this "Natural Stupidity". But they can't help it, they are just a bunch of neurons firing according to simple rules of electrochemistry. There is no such thing as emergence.
Here's one such study that was done showing ai has no reasoning ability: https://ml-site.cdn-apple.com/... [cdn-apple.com]
Re: "AI systems now make decisions" (Score:1)
Umm.. the bias is most probably from the training data so its practically beyond any understanding of the people involved in building or training or running that AI system.
And the bias will occur non deterministically so you can't really prove it very easily.
And even if you identify and prove it, there's not much you can do about it.
So its just the like US justice system.
But if it was like the 2 tier UK justice system then you would be right.
Re: (Score:2)
AI is making decisions for you since you first activated a spam filter that automatically sorts spam into a different folder.
Re: (Score:2)
And Automation != Artificial Intelligence
Re: (Score:2)
Indeed it is. Don't confuse AI with NN.
https://www.researchgate.net/f... [researchgate.net]
Most spam filters are even ML. Even the bayesian filters 20 years ago had a learn spam / learn ham function. And yes, bayesian filters are ML.
Well, the purpose sounds good. (Score:2)
My question is "How will they implement it?". And a secondary question of "Is that what they're really going to attempt?".
Re: (Score:2)
They might be supporting this to erect barriers of entry for new and smaller players that might threaten them.
Natural Justice (Score:2)
This is meaningless (Score:2)
There's no point building policy around the behavior of a system you can't constrain the outputs of.
Even if they succeed in convincing themselves, it's meaningless. A little like banning something people can make themselves without your intervention, except the people are automated processes that spit out pseudo-stochastic data.